Fig 2: Generation of a KI mouse model bearing a pre-rearranged PGT121 ? Strategy for the insertion of PGT121 pre-rearranged VJ into the mouse Ig ? locus. Targeting DNA donor with 5' (3.9 kb) and 3' (3.9 kb) homology arms to the C57BL/6 WT mouse Ig ? locus, murine promoter, leader, and the human PGT121 light chain VJ sequences are located between two homology arms. Two sgRNAs, 18 and 11, were targeted at J4-J5 region of Ig ? locus. CRISPR/Cas9-mediated HDR leads to the insertion of the promoter and PGT121 sequences into the C57BL/6 mouse genome. V segments, enhancers, and the kappa constant regions are shown in gray and labeled appropriately. Yellow rectangles represent J segments; dark blue oval represents the V?4-53 promoter (P); light blue line represents the inserted segment and red rectangles show the rearranged PGT121 VJ. “T” represents TaqMan probe. WT probes were used for the detection of WT allele, Leader probes were used for the detection of the 5'end of the insertion, the specific probes were used for the detection of pre-arranged VJ insertion for PGT121 (probe sequences, see Appendix Table S1).A fragment of genomic DNA (2.2 kb) was amplified by PCR and in vitro sgRNA-guided Cas9-mediated cleavage assay was performed with each of the sgRNAs. sgRNA-targeting sites are indicated by arrows, genomic DNA size is indicated by asterisk.B220+ single B cells from peripheral blood of three PGT121 LC KI naïve mice were sorted. B220+ B-cell populations and their frequencies are shown in FACS plots (left panel). Ig light chains from single-cell sorted B cells were PCR amplified and sequenced. The resulting IGLV libraries were compared to the PGT121 LC reference sequence. The pie charts indicate the frequency of IGLV sequences identical to human PGT121 (red) and mouse IGLV (gray).

Fig 3: Characterization of the specificity, self-reactivity, and phenotype of a collection of monoclonal T cells. (A) Jurkat CD8+ OT-I TCRß cells expressing indicated TCRa clones were activated by T2-Kb cells loaded with indicated concentrations of OVA peptide overnight. The level of CD69 was measured by flow cytometry. Curves were fit using nonlinear regression. (B–E) Peripheral lymphocytes were isolated from sub-lethally irradiated Ly5.1 mice transplanted with immortalized hematopoietic stem cells transduced with indicated clones and the recipient Ly5.2 CD8+ T cells were analyzed by flow cytometry. (B) Expression of the surface marker CD5 on LN T cells. A representative experiment out of three in total. (C) Left – staining of LN CD8+ T cells with fluorescently labeled Kb-OVA tetramer. A representative experiment out of three in total. Right – staining with a dilution series of fluorescently labeled Kb-OVA tetramer. Tetramer binding was measured by flow cytometry. Curves were fit using nonlinear regression. Averages from three independent experiments. (D) Representative FACS plots showing the percentage of AIMT (CD44high CD49dlow) T cells among CD8+ splenocytes from each clone. (E) Representative FACS plots of ISAGhi T cells (CD44low BST2+) and AIMT cells (CD44high CD49dlow) from each clone from mLNs. (F) Quantification of data from three independent experiments. Clone 1: n = 4, Clone 2: n = 4, Clone 12: n = 3, Clone 7: n = 4, Clone 8: n = 3, Clone 17: n = 3. Median is shown. The statistical analysis was performed using two-tailed Mann Whiney U test.

Fig 4: LSD1 depletion results in sustained tumor infiltration of CD8+ T cells.a, b Frequencies of tumor-infiltrating CD3+CD4+ T cells, CD3+CD8+ T cells, CD3-CD49b+ natural killer (NK) cells and CD11b+Gr1+ myeloid-derived suppressor cells (MDSCs) among gated CD45+ leukocytes (a) and cell numbers per gram of tumor (b) analyzed by flow cytometry on day 12 and 18 after MC38 tumor implantation in Lsd1f/f and Cd4-Cre+Lsd1f/f mice (Lsd1f/f d12 group, n = 5; Cd4-Cre+Lsd1f/f d12 group, n = 5; Lsd1f/f d18 group, n = 4; Cd4-Cre+Lsd1f/f d18 group, n = 5). c Calculated ratios of intratumoral CD8+ T cell number over MDSCs (n = 5, 5, 4, or 5, respectively). d–f Representative flow plots (d), percentages (e) and cell numbers (f) of CD8+Tetramer+ TILs in implanted MC38-OVA tumors analyzed by flow cytometry on day 18 (Lsd1f/f group, n = 14; Cd4-Cre+Lsd1f/f group, n = 10). g–i Frequencies of Ki-67-expressing CD8+ TILs (g), frequencies (h) and mean fluorescence intensity (MFI, i) of GzmB-expressing CD8+ TILs analyzed by flow cytometry on day 18 after MC38 tumor implantation (n = 4 per group). j IFN-? expression by MC38-OVA tumor-infiltrating CD8+ TILs analyzed by flow cytometry after stimulation with OVA257-264 peptides or PMA/Ionomycin (P/I) in the presence of GolgiPlug (n = 4 per group). Data represent two independent experiments (a–c, g–j) or are pooled from three independent experiments (e, f) and are presented as mean ± SEM (a–j). Each dot on the graphs represents an individual mouse. Sample sizes are as indicated. Statistical significance was determined by two-sided unpaired t test; ns, not significant (a–j). Source data are provided as a Source data file.

Fig 5: Generation of CLK21 human BCR knock-in mouse model Table shows the total number of pups, the number or frequency of human HC-, human LC-, and human BCR KI pups after One-step CRISPR/Cas9 microinjection of CLK21.CLK21 KI mice F0 and F1 generations. Squares represent male mice and circles represent female mice. Upper halves of squares or circles represent Ig ?, and the lower halves represent Ig H, as shown in the schematic. F0 generation mice genotyping results showing 7, 9 and 14 are HCLK21/WT?WT/WT, 1 and 4 are HWT/WT?CLK21/WT, and 2, 3, 6, and 12 are HCLK21/WT?CLK21/WT. Mouse 3 was crossed with WT to obtain 11 F1 progeny: Five F1 mice are HWT/WT ?CLK21/WT, four are HCLK21/WT?CLK21/WT and two are WT.Single-cell sequencing for naïve B cells from heterozygous CLK21 double KI mice. Left column shows the gating strategy for sorting naïve B cells (upper) and eOD-GT8-specific naïve B cells (lower). Right pie charts show the frequency of paired HC and LC sequences among total naïve B cells (upper) and eOD-GT8-specific naïve B cells (lower).Germline-targeting eOD-GT8 binding activity of B cells from WT, HCLK21/WT?WT/WT, HWT/WT ?CLK21/WT), and HCLK21/WT, ?CLK21/WT KI mice. 8-week-old mice were detected by FACS. X and Y axes represent that BCR were stained with eOD-GT8 tetramer conjugated with Alexa Fluor™ 488 and Alexa Fluor™ 647, respectively. Representative dots were gated from Scatter/Singlet/Live (SSL), B220+ IgM+ IgD+.Quantification of eOD-GT8 binding in CLK21 KI mice. X-axis represents F0 (n = 1) and F1 (n = 3) KI animals, and the Y-axis represents the percentage of eOD-GT8 targeting binders in mature (IgM+ IgD+) B cells. Bars indicate mean ± SD from mice in each group.